[ https://issues.apache.org/jira/browse/SPARK-5972?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14338869#comment-14338869 ]
Manoj Kumar commented on SPARK-5972: ------------------------------------ Just to clarify, this is just to prevent recomputation of the error from previous trees in the computeError methods right?. More specifically lines like these. https://github.com/apache/spark/blob/master/mllib/src/main/scala/org/apache/spark/mllib/tree/GradientBoostedTrees.scala#L198 https://github.com/apache/spark/blob/master/mllib/src/main/scala/org/apache/spark/mllib/tree/GradientBoostedTrees.scala#L203 https://github.com/apache/spark/blob/master/mllib/src/main/scala/org/apache/spark/mllib/tree/GradientBoostedTrees.scala#L233 Or is there anything more to this? > Cache residuals for GradientBoostedTrees during training > -------------------------------------------------------- > > Key: SPARK-5972 > URL: https://issues.apache.org/jira/browse/SPARK-5972 > Project: Spark > Issue Type: Improvement > Components: MLlib > Affects Versions: 1.3.0 > Reporter: Joseph K. Bradley > Priority: Minor > > In gradient boosting, the current model's prediction is re-computed for each > training instance on every iteration. The current residual (cumulative > prediction of previously trained trees in the ensemble) should be cached. > That could reduce both computation (only computing the prediction of the most > recently trained tree) and communication (only sending the most recently > trained tree to the workers). -- This message was sent by Atlassian JIRA (v6.3.4#6332) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org